How Could AI Help Combat Climate Change? Let's Count the Ways

Artificial intelligence (AI) could soon help predict climate change-related calamities. 

Google is introducing new extreme heat alerts in Search to help people stay safe during heat waves. It's part of a growing number of AI applications to monitor global warming. 

"AI is a powerful tool for understanding the resilience of transportation, healthcare, water, energy, and communications systems in response to extreme weather," Auroop Ganguly, a professor of civil and environmental engineering at the Institute for Experiential AI at Northeastern University, told Lifewire in an email interview. "We've used network and data science for predictive insights on everything from the impact of climate on crop yields in the Midwest to the resilience of India's railway system and the US airspace system."

How AI Can Help Climate Change

Extreme heat kills nearly 500,000 people annually, and Google says search interest in heat waves is growing. The search giant says that when users search for information on extreme heat, the new feature means they will see details on when a heat wave is predicted to start and end, tips on staying cool, and related health concerns. 

New Google features could also help cities seek ways to prevent "heat islands," which are urban areas that experience higher temperatures due to structures that absorb heat and re-emit it. Google's Chief Sustainability Officer, Kate Brandt, wrote on the company blog. 

"One strategy to help lower temperatures is increasing tree coverage in these areas. Tree Canopy, part of our Environmental Insights Explorer, combines AI and aerial imagery so cities can understand their current tree coverage and better plan urban forestry initiatives," Brandt said. "The City of Austin has already used the tool to prioritize planting trees in vulnerable areas of the city and even used it to help place bus shelters to increase shade."

AI is also useful for understanding the resilience of transportation, healthcare, water, energy, and communications systems in response to extreme weather. "We've used network and data science for predictive insights on everything from the impact of climate on crop yields in the Midwest to the resilience of India's railway system and the US airspace system," Ganguly said. 

AI Weather Prediction

As the climate changes, high-impact weather is becoming more frequent in many places. Amy McGovern, director of the National Science Foundation's AI Institute for Research on Trustworthy AI in Weather, Climate, and Coastal Oceanography (AI2ES), told Lifewire via email. 

"Being able to both understand how these phenomena will change and improve our prediction of the events is critical to improving our climate resilience," she added. "AI provides us new methods to improve our predictions of the climate as well as to improve the predictions of these high-impact events."

AI can be used to improve long-term predictions of how the weather will look in the future, McGovern said. Simulations of the future climate require vast amounts of computing power, and AI is helping the development of hybrid numerical weather prediction models that blend in AI to be faster and less computationally expensive. 

Her team is using AI in weather prediction for real-life applications. Recently, researchers were able to predict coastal fog in the Port of Corpus Christi in Texas to help the shipping industry and foresee cold events to protect endangered sea turtles. 

"In Oklahoma and New York state, we are using highway camera data to create real-time predictions of visibility, improving road safety," McGovern said. "We are also improving the prediction of hail and tornadoes, with an emphasis on producing nowcasts, which are forecasts for both the 0-1 hour range."

In the future, AI could help decipher data collected from weather observations to make better predictions, Ganguly noted. He said that most historical climate data is based on rudimentary observations from the past. Still, thanks to the advances of modern technology, scientists now have access to a massive amount of information, including from satellites, weather radar, and sensors. 

"El Niño, for example, is primarily caused by surface temperature changes in the Pacific Ocean and has a huge impact on drought and flooding in Central and South America," he added. "We've recently used machine learning to predict river flows on the Ganges, Congo, and Amazon from the surface temperature in the sea."